Forecasting macroeconomic variables in data-rich environments

Marcelo C. Medeiros; Gabriel F.R. Vasconcelos

doi:10.1016/j.econlet.2015.11.017

Forecasting macroeconomic variables in data-rich environments

Marcelo C. Medeiros, Gabriel F.R. Vasconcelos

Research output: Contribution to journal › Article › peer-review

Abstract

We show that high-dimensional models produce, on average, smaller forecasting errors for macroeconomic variables when we consider a large set of predictors. Our results showed that a good selection of the adaptive LASSO hyperparameters also reduces forecast errors.

Original language	English (US)
Pages (from-to)	50-52
Number of pages	3
Journal	Economics Letters
Volume	138
DOIs	https://doi.org/10.1016/j.econlet.2015.11.017
State	Published - Jan 1 2016
Externally published	Yes

Keywords

Big data
Forecasting
LASSO
Model selection
Shrinkage

ASJC Scopus subject areas

Finance
Economics and Econometrics

Online availability

10.1016/j.econlet.2015.11.017

Library availability

Discover UIUC Full Text

Cite this

@article{0fdfaf4c501d4b338392a3d81a5b04b8,

title = "Forecasting macroeconomic variables in data-rich environments",

abstract = "We show that high-dimensional models produce, on average, smaller forecasting errors for macroeconomic variables when we consider a large set of predictors. Our results showed that a good selection of the adaptive LASSO hyperparameters also reduces forecast errors.",

keywords = "Big data, Forecasting, LASSO, Model selection, Shrinkage",

author = "Medeiros, {Marcelo C.} and Vasconcelos, {Gabriel F.R.}",

note = "Publisher Copyright: {\textcopyright} 2015 Elsevier B.V.",

year = "2016",

month = jan,

day = "1",

doi = "10.1016/j.econlet.2015.11.017",

language = "English (US)",

volume = "138",

pages = "50--52",

journal = "Economics Letters",

issn = "0165-1765",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - Forecasting macroeconomic variables in data-rich environments

AU - Medeiros, Marcelo C.

AU - Vasconcelos, Gabriel F.R.

PY - 2016/1/1

Y1 - 2016/1/1

N2 - We show that high-dimensional models produce, on average, smaller forecasting errors for macroeconomic variables when we consider a large set of predictors. Our results showed that a good selection of the adaptive LASSO hyperparameters also reduces forecast errors.

AB - We show that high-dimensional models produce, on average, smaller forecasting errors for macroeconomic variables when we consider a large set of predictors. Our results showed that a good selection of the adaptive LASSO hyperparameters also reduces forecast errors.

KW - Big data

KW - Forecasting

KW - LASSO

KW - Model selection

KW - Shrinkage

UR - http://www.scopus.com/inward/record.url?scp=84949497103&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84949497103&partnerID=8YFLogxK

U2 - 10.1016/j.econlet.2015.11.017

DO - 10.1016/j.econlet.2015.11.017

M3 - Article

AN - SCOPUS:84949497103

SN - 0165-1765

VL - 138

SP - 50

EP - 52

JO - Economics Letters

JF - Economics Letters

ER -

Forecasting macroeconomic variables in data-rich environments

Abstract

Keywords

ASJC Scopus subject areas

Online availability

Library availability

Related links

Fingerprint

Cite this